Functional Significance of Marine Microbial Diversity

D. A. Caron and B. B. Ward

"Undesirable changes in ecosystem processes provide one of the most compelling reasons for conserving biodiversity" (McGrady-Steed et al. 1997)

I. Introduction: History and significance of previous discoveries

Most natural pelagic and benthic ecosystems support highly diverse microbial communities of small eukaryotes and prokaryotes. The term "microorganism" employed here is the general term used by the microscopist -- if it can't be seen without a microscope, it's a microorganism. Thus, we must consider bacteria, archaea, and some eukaryotes (phytoplankton, most protozoa and some fungi). The realization of the ubiquity of microbial communities has unfolded over the last century, but most of our present understanding of these assemblages is based on information collected in the last thirty years. One of the overall goals of research in this area is to attain a mechanistic understanding of how biomass and energy are produced and utilized in marine ecosystems. Here we address the more specific question of how microbial diversity affects ecosystem function. Attaining this latter goal presupposes a knowledge of the organisms composing these communities and their ecological roles.

Microscopic algae have long been recognized as the base of most marine food webs, but the realization that small (<20 µm) primary producers contribute very significantly to total organic carbon production has developed as methodology for examining these organisms has improved. Similarly, the traditional role of bacteria as decomposers in the ocean has broadened as biologists have recognized that bacterial biomass constitutes a significant fraction of the living biomass in aquatic ecosystems. Until the 1970s, it was accepted that the number of bacteria in the sea was represented by those that grew on nonselective culture medium (colony forming units, CFU) and was on the order of 100 cells per ml. With the advent of DNA fluorochromes and epifluorescence microscopy, the abundance estimates increased by three orders of magnitude and direct microscope counts became the standard method.

The detection of 105 to 106 cells per ml by epifluorescence microscopy implied that the majority of cells not detected by culture methods were dead (not viable), or that they were alive but their metabolism involved processes and nutritional requirements which we were unable to duplicate in the laboratory. Simply the idea that there were so many of them implied a significant contribution to mineral cycling by small organisms, much greater than would be inferred from the CFU estimates. Early controversy over whether these cells were viable gradually eased as more and more evidence for activity by natural bacterioplankton accumulated (e.g., direct evidence of radiotracer incorporation, flux estimates based on average uptake and growth rates). However, the validity of epifluorescence counts as estimates of the abundance of viable bacterial cells repeatedly has been challenged as artefactual, and these numbers appear to be on the way down again in the 1990s. Regardless of the exact number, though, fundamental questions concerning these assemblages remain: How many of individual cells are the same/different strains and how many different metabolic capabilities are represented by them? Is there redundancy in the physiological capabilities of different species such that dramatic shifts in ecosystem function do not necessarily result from the loss of a species?

Following somewhat on the heels of the realization that bacteria comprise a significant living biomass in aquatic ecosystems, the 1970s witnessed a shift in the classical phytoplankton-copepod-fish food web paradigm to one in which unicellular, eukaryotic consumers, the protozoa, played a greatly expanded ecological role as predators of bacteria and small algae, and in turn served as prey for larger consumers. These ideas were formalized in a now-classical description by Pomeroy in 1974. This redirection of bacterial (and phytoplankton) biomass through protozoan consumers has been colloquially termed "the microbial loop".

The 1980s and 1990s have brought considerable resolution to the basic concept described by Pomeroy. Simulated in situ tracer experiments showed that microbial communities possess the capability to transform/metabolize just about any compound. Bacteria, the major players in these transformations, were shown to be significant both as food for microbial consumers, and in nutrient uptake and remineralization. New methods for measuring bacterial productivity indicated that up to 50% of the total organic carbon produced by photosynthesis may eventually be consumed by heterotrophic bacteria. Bacterivorous (bacteria-eating) protozoa therefore may consume a substantial portion of the energy available within an ecosystem. In addition, the discovery of two major groups of cyanobacteria (the Synechococcus types and the Prochlorococcus types) on the basis of specific fluorescence characteristics, showed that autotrophic bacterial-sized biomass was a significant contribution to the total microbial biomass in surface waters. Because these small prokaryotic primary producers are too small to be captured by most multicellular organisms, protozoa have been recognized as a link in the food web between small primary producers and heterotrophic bacteria.

More recently (late 1980s and 1990s), the potential ecological importance of viruses has been recognized, particularly as a mechanism for affecting species composition of microbial assemblages. Also, there is a growing realization among aquatic microbiologists of the importance of mixed nutrition (mixotrophy) among microbial species, either within a single microorganism (e.g. phototrophic and heterotrophic) or two or more species functioning in concert (symbioses and other microbial consortia). Viral mortality and mixed nutrition are not new discoveries, but only recently has the ecological significance of these behaviors for marine microbial communities been addressed.

Accepting that microbial food webs are ubiquitous in the environment, and that individual microbes might be very numerous, attention shifted to the identities and activities of these microorganisms. What proportion are autotrophic, heterotrophic, or obligate/facultative copiotrophic/oligotrophic? Because culture methods apparently had failed to detect many prokaryotes in the first place, culture-based characterizations gave way to clever radiotracer/incubation/manipulative experimental techniques. These methods provided (and continue to provide) useful information concerning specific physiological processes, but they failed to identify the organisms responsible for these activities. Immunological methods, very specific but limited to culturable types, began to provide information on the distribution and abundance of individual members of autotrophic and heterotrophic members of picoplankton communities. In most studies, however, the individual target strains comprised a small to insignificant fraction of the total community.

Beginning in the mid 1980s, the rRNA revolution in marine bacteriology began. Using universally conserved rRNA sequences as tools to identify and then to pluck out sequences from both cultured organisms and from bulk DNA extracted from environmental samples, marine microbial ecologists began to contribute to the growing rRNA sequence database and to use that database to study marine microbes. While most of these studies have not been quantitative (we have little information on the relative abundance of different types), it became clear that a very large number of different rRNA sequences were represented by organisms in the environment. Extrapolating from culture collections of bacteria, it was possible to discern evolutionary relationships among organisms which had never been cultured.

The ability to pluck genes out of the environment made it possible to return to the environment and assay those genes and their evolutionary relatives, all without culturing the organisms. However, most of these sequence and relationship data were based on ribosomal sequences; only with often unacceptably unconstrained speculation could these relationships be extrapolated to infer what kinds of activities the organisms who possessed the ribosomes might be engaged in in the environment. By skipping the organismal step, we also missed any opportunity to learn about the ecology or physiology of the microbes we studied. Recently, similar methods have begun to be applied to the study of genes directly involved in transformations of biogeochemical significance in the ocean, and thus to make the connection between the processes and the organisms responsible for them.

II. State of the field: accepted knowledge

In a field in which new methods are published every day, and in which those methods have the capacity to change the way we view "facts" (and frequently have done so), it is difficult to summarize meaningfully what most microbial ecologists accept as current truth. We accept that the abundance of microbes in natural waters and sediments is high (on the order of 105-109 bacteria per ml or gram of sediment, 103-106 cyanobacteria, 102-105 eukaryotic algae or protozoa) and most of the cells are probably metabolically viable if not active. The phylogenetic diversity of these microbes in natural systems is high; a wide range of evolutionary histories and metabolic capabilities coexist in environments that might superficially appear too homogenous to support such great diversity.

Much of our current information about the function of marine microbial communities concerns the abundances and trophic activities of broad ecological categories of microorganisms (e.g. the "bacteria", "phytoplankton" or "protozoa"), or processes taking place within specific size classes of organisms (e.g. <2, 2-200, >200 µm). This approach has been necessitated by the overwhelming diversity of microbial communities, as noted above. These studies have firmly established that most energy (organic carbon) production, utilization and remineralization in marine ecosystems is conducted by microorganisms. Marine microbiology is still very much in the age of discovery, however, with respect to understanding the extent of microbial diversity and how this diversity is environmentally regulated or is reflected in ecosystem function; i.e., how to quantify the significance of microbial diversity. There is still no consensus on how well we have characterized the number of microorganismal species present, how many of these species may be metabolically active at any given time (indeed, we do not know if many of them are even alive), how well our culture collections reflect the diversity of physiologies that are present in the ocean, and what the consequences of reducing this biodiversity may be on energy production and utilization. In fact, at this time it would be very difficult to accurately enumerate the number of microbial species in any locale. Because of the tremendous breadth of species (tens of thousands of species of bacteria and protists have been described), comprehensive studies of microbial diversity do not exist for the vast majority of marine environments.

Recent discoveries and related current foci of research:

Species Diversity of Natural Communities: The application of molecular biological approaches for assessing microbial diversity is still at an early stage (particularly for protists), but already these approaches have indicated a wealth of microbial genotypes in natural communities that are not represented in existing culture collections of marine microorganisms. One particularly interesting example is the discovery of uncultured marine Archaea, which have recently been shown to be ubiquitous in seawater but were previously expected to be restricted to more extreme environments. Characterizing this diversity of natural assemblages and investigating the importance of these genotypes to community structure and function is a major focus in microbial ecology and undoubtedly will continue to be so during the next several years. A good example of the importance of species composition is the formation of harmful algal blooms in which one (or a few) species of microscopic algae strongly dominate primary production and the standing stock of phytoplankton biomass in an ecosystem. Disruption of normal ecosystem function, for example through a reduction in biodiversity, may predispose some environments for invasion and dominance by undesirable species.

Most bacterial diversity information has so far been derived from ribosomal gene sequences, and this approach has recently been initiated for eukaryotic microorganisms as a means of complementing morphologically-based identification schemes. For the most part, however, we have yet to make the connection between ribosomal phylogeny and functional genes associated with particular ribosomally defined types. Efforts to make this link, such as those based on cloning large DNA fragments in artificial chromosomes and combining probing methods with metabolic tracers, show promise. Progress using molecular approaches to study diversity of functional genes has been slow off the mark, first due to a dearth of information on genes which encode functions of biogeochemical significance, and second, because functional genes are not highly conserved, implying divergent evolutionary histories and complicated variable environmental control over biogeochemical processes.

Ecological Roles of Microbial Species: Numerous autecological studies of cultured marine bacteria, microalgae and protozoa conducted largely during the last three decades have provided data on the rates of growth, food assimilation and nutrient cycling by these species. Information on nutritional mode and/or food preferences (e.g. phototrophy, heterotrophy) and size-dependent grazing by consumer populations has been incorporated rapidly into conceptual and mathematical models of energy and elemental flow through pelagic and benthic food webs.

To date, tens of thousands of free-living microbial species have been described, yet we understand the physiological capabilities of only a handful of these species. Experimental studies of individual species of marine microorganisms, both in the laboratory and in the field, will continue to be powerful approaches. However, many species are fastidious in laboratory cultures, and strong evidence (see above) that many have not yet been cultured requires that new methods be developed for ecological study of important species or types.

Viruses have the potential to influence community composition through species specific mortality, and this influence extends to biogeochemical transformations, depending on the viral susceptibility of key species. Similarly, variable nutrient requirements may determine which species can persist in different nutrient regimes. The influence of particular trace metal requirements in phytoplankton nutrition has recently been recognized as a key factor in species distributions; research is needed on the trace element requirements of species to fill this information void for bacteria and archaea.

In addition, there is considerable metabolic plasticity among many prokaryotes and small eukaryotes. Resting stages and/or the ability to lower metabolic rate substantially during periods of adverse environmental conditions make it difficult to extrapolate experimental data from species studied in the laboratory to their activities in nature. As a consequence, new methodologies designed to measure the abundance, small scale distribution and metabolic rates of individual microorganismal species in natural communities are beginning to emerge. These methodologies have begun to provide valuable insights on the activities of microbial species, especially when coupled with physiological data obtained from laboratory cultures.

Expanding Culture Collections: Our ability to interpret and predict the activity of microorganisms in nature is based largely on the extrapolation of experimental work conducted on cultured microorganisms in the laboratory. Obviously, these interpretations are only as good as our ability to culture representative organisms in the lab and there is still a great deal of uncertainty as to whether these species represent the numerically-dominant and physiologically-active microorganisms in nature. As described above, new characterizations of microbial diversity are beginning to address this long-standing issue in microbial ecology. While culture-independent techniques (e.g. DNA sequence-based approaches; mRNA analyses) have removed some constraints from the study of unculturable/uncultured organisms, they have not lessened the usefulness of culture collections. Characterization of cultured species of microorganisms places limits on the activities of these species in natural communities, and cultures continue to form the basis of most probe development. Therefore, continued support of culture collections and research into innovative cultivation methods is very important, even in the molecular age.

Pelagic vs. Benthic Ecosystems: Numerous studies of microbial community structure and function have been performed in the plankton. Much less is known about microbial diversity and processes in the benthos. The spatial, chemical and biological complexity of sediment environments makes these ecosystems very difficult to study, and this complexity has deterred many researchers from more actively pursuing benthic microbiology. Nevertheless, microorganisms in sediments typically are present at abundances that are orders of magnitude greater than abundances in an equivalent volume of seawater, and the presence of plant material in many shallow benthic ecosystems allows for a much greater role of fungi than in the water column or deep-sea benthos.

Interestingly, the range of benthic ecosystems in the ocean encompasses conditions that may be most similar to environmental conditions on primitive earth (i.e. anoxic; hyperthermic). These environments may still harbor microbial species that have evolved very little from early prokaryotes and eukaryotes. The study of these species could hold insights into early evolution.

Microhabitats and other Microorganismal Associations: Pelagic environments traditionally are conceptualized as being more homogeneous chemically and physically, and by inference therefore more biologically homogenous than benthic habitats. However, recent studies of the ecology of suspended detrital material (so-called "marine snow" and the more ubiquitous and abundant microscopic aggregates) have indicated more heterogeneity in the plankton than was previously supposed. Individual species have discrete distributions in time and space, such that community organization may influence the rate and distributions of biogeochemical processes that are commonly modeled as bulk reaction rates. Technical advances are making it possible to resolve the spatial and temporal localization of microbes in relation to physical and chemical gradients, in relation to processes in the environment, and in associations among two or more species of microorganisms (symbioses, commensalism, parasitism).

Quantitative methods: A great diversity in rRNA and functional gene sequences implies that a great number of genetically unique microbial types can be found in the environment. However, we know very little about species equitability: that is, whether a community or assemblage is composed of fairly equal numbers of many different types, or if it is dominated functionally by the prevalence of one or a few types. Methods now becoming established and emerging include those that indicate the abundances (and activities) of individual strains or species of microorganisms, as well as `bulk' methods that address the abundance and activities of entire assemblages or communities. These approaches will help define the contributions of specific microorganismal types to the entire microbial milieu. Advances in automated image analysis and flow cytometry should make acquisition of these data more rapid and reproducible.

III. Existing infrastructure

Marine microbiology is a technique-limited endeavor. Our understanding of the role that these assemblages play in aquatic ecosystems has expanded in direct correlation to application of new approaches and methods for studying these populations in nature and in the laboratory. Modern conceptual and methodological approaches have been adapted largely from the fields of classical microbiology (bacteriology, protistology, virology, mycology), and more recently from biomedical research in general. These pathways will continue to be major sources of technological innovation in marine microbiology.

At the same time, effective exploitation of these approaches and methods will continue to be enacted by individuals with strong backgrounds in classical and marine ecology. The melding of these different backgrounds (either as effective collaborations between scientists with complementary backgrounds, or as cross-training of individuals) has contributed valuable insights and methodologies to the study of marine microorganisms. This amalgam (maintenance of the intellectual and technological conduit from classical microbiological and biomedical fields into ocean science, and the application of these approaches to ecological questions) is essential for future progress in marine microbiology.

Unlike medical microbiology, methods for studying microbial assemblages in nature tend to aggregate organisms into large `guilds' and represent them by a generalized or average behavior. Three factors are important in explaining this dichotomy. First, identification of clinical isolates is aided by the fact that relatively few microorganisms cause disease in humans (relative to the total diversity of microorganisms) and there is a large amount of background information on many (perhaps most) pathogenic species of microorganisms. Second, most clinical studies are concerned with target species on a `presence/absence' basis, rather than an absolute number or relative abundance.

Third, much of the reason for these `aggregated' approaches to microbial communities is a result of the necessity (to some degree) to conduct this work within the context of large multi-disciplinary biogeochemical programs. These programs clearly have contributed to our understanding of microbial ecology in the ocean. However, the goals set forth by these programs generally do not foster in-depth studies of individual species, but rather the collective impact of their behavior. While the latter information is useful, it does not promote a mechanistic understanding of how organism-organism interactions lead to the aggregate behavior observed.

Another limitation of microbial ecological research is that most of the variables we think are important to measure are difficult and time-consuming to measure. These issues limit the spatial and temporal coverage that can be attained, especially in light of oceanographic spatial scales.

IV. Exciting future opportunities and challenges

Opening `Black Boxes' (Assessing Microbial Diversity): As indicated above, there is a propensity among aquatic biologists to employ methods that are designed to characterize metabolic activity of an entire assemblage of species in a single measurement (e.g. the measurement of bacterial production by leucine or thymidine incorporation; the measurement of total herbivory by removal of grazers by filtration or dilution). These approaches have been necessary in order to grasp the overall importance of various biological assemblages in broad biogeochemical roles. In particular, comparisons of different methods to constrain a single process has been useful (e.g. simultaneous measurements of phytoplankton population growth and herbivore removal). Without question, however, these `community level' measurements mask the variability within the biological community. More importantly, they do not typically provide a mechanistic understanding of the effects of shifts in community composition on ecosystem function.

At present we know very little about the species diversity of most natural microbial communities. Establishing the breadth of microorganismal species diversity (and therefore potential physiologies) is a fundamental goal of marine microbiology for the next few decades. The complementation of classical methods of studying these communities (culture, microscopy) with modern molecular biological approaches and technology will provide powerful new methods for addressing this issue.

In situ Activity: Establishing the presence (and abundance) of species in a microbial community is an important step in assessing biodiversity. Establishing activity is a separate but equally important matter. It is likely that only a fraction of the assemblage is metabolically active at any given time, but the proportion of the community that is active (and which species) is still largely undetermined. Small eukaryotes and prokaryotes are capable of extremely rapid growth rates (doubling their number several times per day), but many are capable of existing in a slowly-growing or non-growing state while still remaining viable. For this reason, there is still considerable controversy as to the functionality of much of the microbial biomass in the ocean. The development and application of methods for measuring the in situ metabolic rates of individual species of microorganisms is essential for future studies in microbial ecology. Miniaturization and automation of sensor technologies (e.g., chemical and molecular probes on microchips) will probably play a role in this research, especially if we are to address the need for more comprehensive spatial and temporal data coverage for microbiological variables.

Non-traditional trophic behaviors: Most research involving organism-organism interactions in the ocean has concentrated on predator-prey interactions. We must begin to characterize other interactions including the behaviors of symbiosis, commensalism and parasitism, and begin to develop an understanding of how these relationships affect microbial community structure and function. Bacterial-metazoan symbiosis has moved to center stage in deep-sea biology during the last two decades with the realization that hydrothermal vent communities are sustained by sulfide-oxidizing endosymbiotic bacteria. Research on marine photosymbioses has been relegated largely to metazoan invertebrate-zooxanthellae systems. Nevertheless, these kinds of relationships abound within and between microbial phyla, and the ecological/biogeochemical significance of these relationships is only beginning to emerge. Molecular biological approaches will increasingly be used to complement `whole organism' approaches. This combination holds the potential for revealing the underlying molecular bases of symbiotic interactions; that is, the specific biochemical signals that are responsible for establishing and maintaining these relationships and their genetic bases.

V. Summary

Much of our present understanding of marine microbial diversity involves the abundance of broad groups of microorganisms and their overall biogeochemical activities in some specific environmental settings. We need to move beyond this `black box' approach in the future if we are to realize a better understanding of community function and the role of specific microbiological entities within these communities. Clearly, this is a long-term goal given the tremendous diversity that exists within microbial communities. It is not realistic to expect that all microbial species can be characterized, nor is it realistic to expect that even if this goal could be achieved that the function of marine ecosystems would be completely understandable or neatly predictable. Nonetheless, characterization of the breadth and meaning of marine microbial biodiversity will undoubtedly reveal much about the manner in which these communities are structured and how they function. It will also yield fundamental information on the evolution of biogeochemical cycles and their regulation on micro to global scales. Biotechnological advances and applications arising from this knowledge will be a huge added bonus. Anticipated discoveries concerning the activities of these species make microbial ecology an exciting pursuit at the cutting edge of biological oceanography.

VI. Citations

Fuhrman, J. A., Suttle, C. A. (1993). Viruses in marine planktonic systems. Oceanogr. 6: 51-63.

Jannasch, H. W., Taylor, C. D. (1984). Deep sea microbiology. Ann. Rev. Microbiol. 38: 487-514.

McGrady-Steed, J., Harris, P. M., Morin, P. J. (1997). Biodiversity regulates ecosystem predictability. Nature. 390: 162-165.

Pace, N. R. (1996) New perspective on the natural microbial world: Molecular microbial ecology. ASM News. 62: 463-470.

Pomeroy, L. R. (1974). The ocean's food web, a changing paradigm. Bioscience. 24: 499-504.

Wiebe, W. J., Moran, M. A., Hodson, R. E. (ed.) (1994). The microbial loop. vol. 28(2). Springer-Verlag New York, Inc., New York.

Return to OEUVRE Topics Page

National Science Foundation